(cherry picked from commit bddb2cd3)

Train with pipe reader merge develop

Support imperative learning rate scheduler

* Add linear learning warmup method

This warmup lr can be combinated with other learning rate strategies.
For example:
            decayed_lr = fluid.layers.linear_lr_warmup(
                fluid.layers.piecewise_decay(boundaries, lr_steps),
                warmup_steps, start_lr, end_lr)

* fix cudnn batch norm accuracy test=develop

* fix cudnn batch norm accuracy test=develop

* disable failed test for later fix test=develop

Fix document array_read code error

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

remove the useless check 

* fix compiled
test=develop

* follow comments test=develop

test=develop

KL div loss: add kldiv_loss op

Add attr 'axis' for softmax

Add temporal_shift op for TSM model

* Add DecoupledWeightDecay

remove profiling from int8 analyzer test

test=develop

test=develop

test=develop

refine subgraph trt and anakin

* add slice support for dim < 0 test=develop

test=develop

preprocess with PIL the full val dataset and save binary

create test for quantized resnet50

test=develop

test=develop

test=develop

* fix comments of 16410, test=develop

* modify inplace_op_inference_test according to pass interface change, test=develop

test=develop

test=develop

test=develop

[slim]  Fix checkpoint of quantization strategy.

* Add a tool to summary model's PARAMS, FLOPs in paddle/fluid/contrib.

test=develop